To delineate the joint roles of genetic predisposition and radiation exposure in the etiology of second primary breast cancer, we propose to apply a genome-wide association (GWA) approach to women with bilateral breast cancer for whom we have detailed radiation dose estimates. The existing study population consists of 708 women with asynchronous bilateral breast cancer (cases) and 1,399 women with unilateral breast cancer (matched controls). In addition, through four cancer registries in the US and one in Denmark, we will recruit, interview and obtain a blood sample from 800 women with bilateral breast cancer (new cases) and 800 women with unilateral breast cancer (new controls). Our Specific Aims are as follows: AIM 1: Identify SNPs that are associated with bilateral breast cancer using a two-stage approach in a population-based case-control study. In Stage 1 for Discovery, perform a GWA study of bilateral and unilateral breast cancers to identify common SNPs associated with the incidence of bilateral breast cancer using the existing cases and controls. In Stage 2 for Validation, evaluate the 6,000 most significant SNPs identified in Stage 1 in the new cases and new controls. AIM 2: Identify SNPs that appear to interact with radiation exposure. Using the same two-stage design as for Aim 1, incorporate detailed information on radiation exposure to discover and validate the 6,000 most significant common SNPs that interact with radiation exposure and modify the risk of developing a second primary breast cancer. AIM 3: Using the entire WECARE Study population, determine whether genomic regions found to be associated with unilateral breast cancer as identified via independent GWA studies are also associated with risk of developing bilateral breast cancer, or radiation-induced breast cancer. AIM 4: Characterize genomic regions containing variants associated with bilateral breast cancer and/or radiation-induced breast cancer in a population-based case-control study design. The 10 most strongly implicated regions associated with potential causal SNPs identified in Aims 1, 2, or 3, will each be further characterized by re-sequencing to identify other variants for finer mapping of these regions. We will genotype putative disease associated and/or functional genetic variants in our entire study population of 3,707 women. Results from this study will help us understand genetic and environmental factors that influence susceptibility to breast cancer in particular and radiogenic cancers in general.